The Brewing Legacy Systems Storm
There is a category 5 storm brewing in the IT industry which will affect all sectors of the economy. Some corporate executives are aware of the operational risk coming at them in the next few years, but most do not realize its full extent. If IT management thinking does not change quickly, many large corporations will be too far down a path to destruction to ever reverse direction.
Several forces are now coming together simultaneously which may lead to a world-wide collapse in critical business systems operations:
- The struggling economy is causing huge cutbacks in IT professionals, salaries, training, and equipment spending. The legacy mainframe systems workforce is nearing retirement and is not being backfilled. Younger IT workers are not being trained by the older, experienced professionals due to hiring freezes.
- Many companies are sending an internal message that their core mainframe systems will be moving to distributed to commodity chip platforms using ‘newer’ technologies, causing younger workers to shy away from mainframe careers.
- Distributed smaller systems, having a lower cost of acquisition, are falsely perceived to have a lower total cost of ownership (NYSE:TCO). The true future people, energy, and software costs of supporting applications and data on thousands of servers will cripple some businesses.
- Distributed computing systems will lead to much more complex networked architectures. They will be much more difficult to diagnose performance problems, and to recover in the event of a failure or site disaster.
- Company loyalty and longevity in the job has all but disappeared due to the abandonment of defined pension plans and other benefits reductions. Transferable 401K plans will mean that as soon as skills are acquired, IT professionals will be looking for better opportunities.
- Company executives are doing a terrible job of managing critical IT expertise through expense cuts, salary reductions, and layoffs. Aging experienced IT professionals, in declining health and approaching retirement, and with heavier workloads being heaped on them, are loosing their enthusiasm for the business.
One estimate states that US corporations have over a trillion dollars invested in legacy systems that run their businesses. Another estimate is that there are close to 400 billion lines of legacy third generation language (3GL) code, like COBOL and PL/I, as well as a great deal of 2GL code like Assembler. These systems touch each of our lives every day. It would be impossible to book a flight, car, or hotel, or even cash a check or use an ATM without them. They run all sectors of our economy. They are our silent and occult crumbling infrastructure.
Most of these systems have little documentation other than the knowledge that exists in the heads of the soon-retiring IT professionals that support and run them today. All of the business rules that have been built and evolved over the past 50 odd years are imbedded in them. This means that as the current workforce retires and younger workers are not trained to replace them, critical systems will become quite difficult to support or recover.
The skills problem is already happening. I have personally gone in to companies whose systems were down to the point where they were out of business with no hope of recovery on their own. I have personally recovered their databases and gotten their applications back up and running. The last one was a major insurance carrier who had lost their claims database. They should have had in-house skills to avert the problem in the first place, and to recover from the problem themselves, which by the way they caused. But they did not due to attrition and lack of backfill. The scary part is that if we had not fixed the problem within 72 hours, it would not have mattered because they could have never caught up in their processing due to of lack of machine capacity as a result of their tight budget.
Throwing money and people at the problem after it gets here is not the answer. Anyone who has read The Mythical Man Month http://en.wikipedia.or...
understands the fallacy in this approach. Legacy systems can be very large and extremely complex. The knowledge required to support them has been acquired over decades. That knowledge will soon disappear in a matter of a few years, almost as if an IT dark ages has befallen us. When this happens and critical systems begin to fail, the economy is bound to plunge into an abyss with no bottom in sight. Imagine not being able to get at money to buy essentials like food, clothing or shelter, or to produce and ship those essentials in the first place. Imagine not being able to conduct financial transactions with banks, brokerages, insurance carriers, retail stores, travel and transportation, or gas stations to name a few. We live in a global wired economy, and when the wires go silent the economy ends and anarchy ensues.
Corporations must act now to shadow older IT professionals with younger workers. They must make major investments in legacy systems skills and platforms like the mainframe. They must send a message that their legacy systems will be running their businesses far into the future and that IT careers in these systems are viable and secure. Finally they must show their employees, in real terms, that they value their IT staff. Instead of creating temporary make-work government jobs, like census taking, our government should be stimulating the economy by helping corporations to bear the cost of legacy skills investment which will create permanent jobs that pay taxes.
IBM is doing their part to get deliver the message and get out in front of the problem. Several years ago IBM launched its Academic Initiative http://www.istockanaly...
and have partnered with over 400 schools and universities to bolster mainframe skills. IBM has also put much of the coursework on-line, all for free. In addition IBM has completely modernized the mainframe to the point where it may be the most open application and database consolidation server available, running over 8 operating systems including several flavors on Linux. Mantissa is opening up the mainframe to run MS/Windows servers under its z/VOS operating system. But there is only so much IBM can do.
Mainframe platform skills are only a small part of the equation. The greatest challenge facing companies is coming to a realization that the mainframe may be the only platform that can support many of their core business operational requirements, and their willingness to invest in hiring and training younger workers to learn and support their applications built on mainframe capabilities. Finally, they must be willing to deploy new applications and exploit the new capabilities of the modern mainframe.
Many may call me an alarmist. They will say that we will find a way out of the coming crisis, if it happens at all. I do not think so. Large businesses must begin to act now in order to avert the crisis, for when it does arrive it will be too late.
As far as investment advice, take a look at IBM’s 5 year chart and you will understand how strong a company IBM versus its computing hardware competitors. It closed above $105 yesterday, not that far from its high of about $130 in late July 2008. IBM is not immune from the downturn in the economy, but it has fared far better recently than most other large companies.
Disclosure: Stan Muse is LONG on IBM.